linear-claude-skill

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill's MCP configuration explicitly passes the runtime endpoint "https://mcp.linear.app/sse" to the mcp-remote command, which is invoked at runtime to perform remote tool operations (i.e., execute commands/operations on a remote server), so this external URL can execute remote code that controls agent behavior.